Rapa Nui Journal: Journal of the Easter Island Foundation

Volume 21 Issue 2 October Article 2

2007

Chronology, deforestation, and "Collapse:" Evidence vs. faith in Rapa Nui prehistory

Terry L. Hunt University of Hawai'i

Carl P. Lipo California State University, Long Beach

Follow this and additional works at: https://kahualike.manoa.hawaii.edu/rnj

Part of the History of the Pacific Islands Commons, and the Pacific Islands Languages and Societies Commons

Recommended Citation Hunt, Terry L. and Lipo, Carl P. (2007) "Chronology, deforestation, and "Collapse:" Evidence vs. faith in Rapa Nui prehistory," Rapa Nui Journal: Journal of the Easter Island Foundation: Vol. 21 : Iss. 2 , Article 2. Available at: https://kahualike.manoa.hawaii.edu/rnj/vol21/iss2/2

This Research Paper is brought to you for free and open access by the University of Hawai`i Press at Kahualike. It has been accepted for inclusion in Rapa Nui Journal: Journal of the Easter Island Foundation by an authorized editor of Kahualike. For more information, please contact [email protected]. Hunt and Lipo: Chronology, deforestation, and "Collapse

Chronology, deforestation, and "collapse:" Evidence vs. faith in Rapa Nui prehistory

Terry L. Hunt Department ofAnthropology, University ofHawai'i

Carl P. Lipo Department of Anthropology and IIRMES, California State University Long Beach

n recent publications (Hunt and Lipo 2006; Hunt 2006, concerns about our own environmental future, Rapa Nui I 2007) we have presented a detailed and comprehensive offered the quintessential case of "ecocide," as Jared Dia­ analysis of new and existing archaeological inforn1ation as mond (2005) dubbed it. The case for "ecocide" seemed con­ it relates to the date of Rapa Nui's colonization, the island's sistent with some accounts from early European visitors, ecological transformation and the assumed relationship to some of the oral traditions, Heyerdahl's views of pervasive "collapse." After reviewing published dates and our results warfare and cultural replacement, and the emerging palaeo­ at 'Anakena, we came to the conclusion that although it is ecological evidence. Rapa Nui provided a compelling tory conceptually possible that humans arrived on the island and environmental message that held relevance in today's many hundreds of years prior to AD 1200 , there is cur­ urgent global crisis (e.g., Kirch 1997,2004). While we are rently no empirical support for believing this wa so. Until certainly concerned about contemporary environmental unequivocal evidence emerges for earlier colonization, our problems, we a k, have the cau es and consequences of understanding of the island's prehistory mu t be founded on Rapa Nui deforestation been misconstrued in accounts for a shorter chronology ofabout 800 years. "ecocide" and cultural collapse? Our re-examination of the evidence for Rapa Nui has Recent field research has changed some of the details provoked quite negative reactions (Flenley and Bahn 2007a; of Rapa Nui's prehistory (e.g., Hunt and Lipo 2006; Hunt 2007b). In this paper we address some of the issues raised 2006, 2007), and received widespread pres (e.g., Gibbons in the responses to our work. We argue the critical problem 2006; Lovgren 2006;Young 2006). A series of eight new has come from ignoring the empirical sufficiency of ar­ radiocarbon dates from excavations of stratified deposits at chaeological and palaeo-environmental conclusions. Unfor­ 'Anakena show occupation began by approximately AD tunately, the popular narrative, one of more than a millen­ 1200. nium ofpopulation growth resulting in environmental catas­ The shorter chronology for Rapa Nui follows the pat­ trophe and demographic collapse, so often repeated for tern that has emerged for chronologies throughout eastern Rapa Nui, has enjoyed precedent over the actual evidence, Polynesia (e.g., Anderson 1991; Anderson and Sinoto 2002; or lack of it. Contrary to the strictures of science, continued Athens et al. 2002; Burney and Burney 2003; Higham and faith in a particular story for Rapa Nui has closed the minds Hogg 1997; Kennett et al. 2006; Spriggs and Anderson of orne scholar to considering new evidence, or re­ 1993). Moreover, recent research on deforestation and avian evaluating the old. extinctions in the Hawaiian Islands (Athens et al. 2002) has In a remarkable set of discoverie , John Flenley and demonstrated that the Pacific rat, introduced to islands by his colleagues (e.g., Dransfield et al. 1984; Flenley and Polynesian colonists, irrupted into millions, expanding over King 1984; Flenley et al. 1991) found that the pollen evi­ the landscape many times faster than people (see Fenchel dence from a series of lake cores indicated that Rapa Nui 1974). This invasive species consumed seeds of native once supported a forest dominated by a giant palm tree (an plants, effectively depressing or halting the regeneration of extinct Jubaea sp. [Grau 2004] or Paschalococo disperta many forest taxa. The ecological impacts of rats are well [Dransfield et al. 1984]). Their work confirmed what others documented (e.g., Campbell 1978; Campbell and Atkinson had suspected about a native forest that bad disappeared 1999, 2002; Town et al. 2006). Before the effects of fire, (e.g., Skottsberg 1956; Mulloy and Figueroa 1978:22). felling, or other direct human actions, Athens et al. (2002) Flenley's pioneering palynological work showed that Rapa have shown that the introduced rat de troyed the Nui had undergone a dramatic ecological transformation Pritchardia palm forests of the 'Ewa Plain of O'ahu Island. with deforestation and associated extinctions (Steadman et Other palaeo-environmental field studies (e.g., Athens al. 1994; Steadman 2006). 1997; Denham et al. 1999) point to comparable impacts on It seemed obvious to researchers that Rapa Nui was a native forests from rats. The implication is that rats played a clear case of human recklessness, over-population, over­ significant role in the collapse of forest over much of the exploitation, and cultural collapse. Given contemporary lowland areas of the Hawaiian Islands as well as elsewhere

This paper has been peer reviewed. Received 14 March; Accepted 13 April 2007; Revised 14 May 2007

Rapa Nui Journal 85 Vol. 21 (2) October 2007

Published by Kahualike, 2007 1 Rapa Nui Journal: Journal of the Easter Island Foundation, Vol. 21 [2007], Iss. 2, Art. 2

in the Pacific. If rat alone devastated huge tracts of Hawai­ ian forest, what relative impact might they have had on A LONGER OR SHORTER CHRO OLOGV? Rapa ui, an island a fraction of the size with a depauper­ ate biota, and no natural predators (Figure I)? This question New Evidence is partially answered by considering the relatively simple, Renewed excavations of the stratified deposits on the but almost ideal "rat-fodder" native vegetation of remote north ( eaward) side of Ahu au au at 'Anakena Beach Rapa ui: the ancient palm and its abundant nuts (Hunt have yielded significant evidence for the chronology of the 2007). island's colonization (Hunt and Lipo 2006; Hunt 2007). Long held to be the site of early ettlement (e.g., Routledge 1919:241-242; Heyerdahl 1961 a:34-36; Steadman et al. 1994; Skj01svold 1994a), 'Anakena is the ite of a tratified dune with cultural depo it extending down to a natural clay ) substrate. This clay substrate contains an organically en­ ( riched pa1aeosol with abundant cultural materials (e.g., ob­ ~Ir sidian artifacts, rat bone , charcoal) embedded in it upper­ most 5-10 cm. Below thi ancient horizon is a natural undi ­ turbed clay deposit riddled with the root molds of the ex­ tinct palm. The root mold provide evidence that natural sedimentation (weathered volcanic a h) occurred at orne ancient (geologic) time, and had been undi turbed for at lea t several centuries before AD 1200. Our three ea ons of excavations and deep coring have confinned that thi basal clay substrate extends several meters below and is entirely devoid of cultural materials. Thus, the palaeo 01 at 16 Kilometers the interface of the clay substrate and the sand layers above L...J...... L....Jc...J...... t...... L.....J'-" fonned a stable ground surface at the time of first coloniza­ tion of the island. Thi tratigraphic interface provide a Figure I. caled overlay of Rapa Nui and O'ahu island; com­ sound context to date the arrival of Polyne ian coloni t on pare the relatively large area of Pritchardia deforestation of the Rapa Nui. Multiple radiocarbon dates from thi context and 'Ewa Plain (and lowland O'ahu generally) with the small total 2 the sand dune abo e how a consistently ordered chronol­ size ofRapa Nui (171 km ). ogy beginning around AD 1200. Considering the number of radiocarbon dates, consistency of chronological results or­ dered stratigraphically, and analysis of the geomorphic and REACTIO S edimentary context, these re ults from the 'Anakena Dune are among the mo t complete published for Rapa ui. In a letter to the editor of American Scienti t and the paper recently publi hed in this journal, Flenley and Bahn Veracity ofEarliest Cultural Deposits (2007a, 2007b) que tion the legitimacy of new evidence Flenley and Bahn (2007a, 2007b) claim that our evi­ and it implications for Rapa ui prehi tory (Hunt 2006, dence for chronology i weak. They imply a ert that no 2007). They appear to simply defend views previously pub­ natural (i.e., pre-cultural) layers are beneath the cultural Ii hed (e.g., Flenley and Bahn 2002; ee also Diamond ones that we excavated. They also assert that earlier cultural 2005). Flenley and Bahn (2007a, 2007b) assert that ar­ layers could have existed but are now missing. Flenley and chaeological and botanical data unambiguou Iy point to Bahn (2007a, 2007b) uggest that older layers of sand human over-exploitation of resources on Rapa Nui, and that might have blown or washed away before those we exca­ thi interpretation i consistent with an account that centers vated were deposited. Sand dunes are dynamic environ­ on elf-induced collapse or Diamond' "ecocide" prior to ment ; they develop, erode, and change rapidly. Yet the un­ European contact. They challenge our primary evidence and di turbed state of the original in situ clay ub trate (i.e., careful examination of the archaeological and palaeo­ below the and layer) i indicated by artifact , charcoal, rat environmental record , and instead lobby for the accuracy bone, etc., embedded in a palaeosol directly above a oci­ of their tory for the island. Here we argue that it is critical ated ancient palm root molds. Moreover, the radiocarbon that we evaluate the evidence, not ju t cling to long-held chronology of the and dune layers above reveals a strati­ beliefs ba ed on faith in the absence of evidence, problem­ graphically-con istent chronology for rapid sand dune de­ atic oral hi tories, and appeals to authority concerning the velopment. While the notion of "missing deposit" i con­ i land' ecological and cultural past. ceivable, they are nonetheless not there. Our excavation and deep probing at 'Anakena revealed the earlie t sign of human presence, with no evidence in tIle clay depo it be-

Rapa ui Journal 86 Vol. 21 (2) October 2007 https://kahualike.manoa.hawaii.edu/rnj/vol21/iss2/2 2 Hunt and Lipo: Chronology, deforestation, and "Collapse

low. We maintain that, until unequivocal indications emerge Population Growth (3%) for earlier cultural strata, claims of "mi ing deposits" are ju t special case pleadings for negative evidence. 4000 -, 3500 I c 3000 ~ 1 Veracity of 'Anakena Deposit a Evidence for 2500 / -3 2000 / Colonization ~ 1500 In addition to que tioning our chronology for / IL 1000 ..7 'Anakena, Flenley and Bahn (2007a, 2007b) challenge 500 whether we can tate that the date obtained there reflect o ---' initial or early colonization. They ask, if 'Anakena was the 1100 1200 1300 1400 1500 site of the fLfst settlement, why are there settlements at two Year AD other locations at the same date? Of course, there is signifi­ Figure 2. Plot of human population growth on Rapa ui ba ed on cant ambiguity a to what the" arne date" mean in hi tori­ an initial logistic rate of3.0% (see Bird ell 1957 where he docu­ cal per pective. The" arne date" in the measurement tem1 ments three cases of logistic growth in human populations as high of radiocarbon dating is roughly within a century or about a 3.4%). [fwe assume a small founding population ofabout 50 five human generations. There would be plenty of time for arriving on Rapa Nui in AD 1200 with an initially large growth even a small number of people - certainly those compri ing rate enabled by an unpopulated landscape, this rate, in one hun­ five generations - to travel 10 or 20 kilometer to other dred year, would produce a population ofapproximately 1200. umbers consistent with earliest ob ervers (e.g., 3-4000) could part of the island. There are no natural barrier to stop peo­ have easily been reached before AD 1400, even considering that ple from venturing acros the island regardless of where the growth rate would have slowed a population filled the i land. they fir t ettled. We would even expect to find structures built' at the arne date" regardle of the years or decade between their con truction event. Given the island' ize, colonization and human expan ion over the island should Independent Evidence ofFirst Colonization appear contemporaneously as a con equence of the preci­ Flenley and Bahn (2007a) imply in their argument that sion ofarchaeological dating. Polyne ian coloni t fLf t et foot el ewhere on the i land, thu the later (ca. AD 1200) chronology now established at As umption ofSlow Population Growth Rates 'Anakena repre ent a ignificantly later expansion to thi Flenley and Bahn (2007a, 2007b) make a tacit a ­ prime part of the i land. We find such an idea dubiou given umption that Rapa ui experienced uncharacteristically the island's mall size and the probability of rapid human low population growth, despite the speed with which the population growth. But there is another set of evidence that re t of ea tern Polynesia was ettled. If one were to assume points to the event ofcolonization at about AD 1200. While a low population growth rate (e.g., ca. 1%) we might imag­ Flenley and Bahn might uppo e Polyne ian coloni ts ine a scenario in which centuries of Polynesian presence on waited centurie after their arrival to leave traces of their Rapa ui would remain all but invisible. Like the assertion pre ence at 'Anakena, it i difficult to maintain the idea that of "mi ing deposits," uch a cenario is based on the lack introduced Pacific rat also waited centuries before their of evidence (see Ander on 1995 on the notion of "cryptic expansion to 'Anakena (e.g., ee Fenchel 1974; Wilm hurst settlement" in East Polynesia). The few documented cases and Higham 2004 on rat expansion and human colonization of an isolated group of humans colonizing empty environ­ in the considerably larger islands of New Zealand). The ments point to population growth rates of around 3.4% evidence is unambiguous: at 'Anakena, rat bones are com­ (Bird ell 1957). Based on a comparable rate, we would ex­ pletely absent and then first appear in great abundance in pect that even a small number of people (50) would expand the same basal layer with charcoal, ob idian, other artifact , to well over a thousand in ju t over a century (Figure 2). and other faunal remains of fi h, bird, and ea mammal Specifically, population could grow to 2000 in ju t (Hunt 2007; see also Steadman et al. 1994, Skjl'Jlsvold 123 year starting with 50 or in 100 years from 100, in 76 1994b). 0 cultural remain pre-date, in the stratigraphic or year from 200, and in 63 year from 300. Even the smaller chronological record , the pre ence of the introduced rat. tarting populations can grow to significant numbers in a hort time. In light of documented human population Rejection ojStandard MeansJor Evaluating Ra­ growth rates, it seems some researcher in the Pacific are diocarbon Dates held in the embrace of a contradiction: rapid population Finally, we find it di heartening that Flenley and Bahn growth enabled colonization of hundred of far-flung i ­ (2007b: II, emphasis added) regard the careful evaluation of land almo t imultaneou Iy but then uddenly growth re­ the bridging argument between radiocarbon date (Figure verted to a remarkably, indeed inexplicably, slow rate that 3 and 4) and archaeological conclu ion a merely follows once these colonist settled their newly discovered "fa hionable." Seeking reliable and valid an wer employ­ i land. Such a contradictory argument violates the basic ing an empirical tandard - the ba i of cience - is not tenet ofdemography and evolutionary biology, and ignifi­ merely fashion. We analyzed radiocarbon date to under- cantly, the actual documented cases (Birdsell 1957).

Rapa Nui Journal 87 Vol. 21 (2) October 2007

Published by Kahualike, 2007 3 Rapa Nui Journal: Journal of the Easter Island Foundation, Vol. 21 [2007], Iss. 2, Art. 2

Calibrated Age Ranges

U Il

ann­ ann­ ~DD ann­ ann­ ann· ann­ ann· ann­ ann· D ~~~:l ann- I~ ann· o ann· nn· ann­ ~OD ann­ ann· ann­ [I]!IJ [=-=::IJ ann· ann· D ann­ ann­ [I] ann-

1000. 1100. 1200. 1300. 1400. 1500. 1600. 1700. 1800. 1900. 2000.

cal AD Figure 3. Calibrated radiocarbon dates (n=42) for charcoal as ociated with deforestation and erosion of primeval oils from location throughout Rapa Nui (data from Mann et al. 2003: 135; Mieth and Bork 2004, 2005; Mieth et al. 2002).

tand what we know reliably about the archaeological re­ replicated by one or more samples from the ame or directly cord on Rapa Nui. To accompli h this we u ed a widely adjacent stratigraphic context. Replicating date, while not accepted protocol known as "chronometric hygiene." As foolproof, reduces the probability of accepting re ult that pointed out some time ago (e.g., Dean 1978; Dunnell and may be from samples of high-inbuilt age, particularly where Readhead 1988), radiocarbon date cannot alway be unam­ the wood taxon has not been identified. A the exten ive u e biguously related to an event of archaeological interest. In of "chronometric hygiene" in New Zealand (e.g., Anderson this way the "chronometric hygiene" approach provides 1991; Higham and Hogg 1997) and elsewhere in the Pacific simple guidelines for accepting dates as reliable and valid (e.g., Spriggs and Anderson 1993; Liston 2005) has demon­ measures of the prehistoric event we investigate. strated, there are many ways in which radiocarbon dates For example, the protocol tandardize efforts to avoid (i.e., the radiocarbon event) can be significantly older than the common problem of high-inbuilt age or other purious the archaeological event (e.g., colonization, occupation, re ult from samples of wood/charcoal from long-lived etc.). tree, ample of mixed materials (e.g., charcoal and oil), In our work, the radiocarbon date excluded in e ti­ ample for which corrections for isotopic fractionation, mating the chronology for Rapa ui's colonization were reservoir effects, etc., have not been made, or samples not

Rapa Nui Journal 88 Vol. 21 (2) October 2007 https://kahualike.manoa.hawaii.edu/rnj/vol21/iss2/2 4 Hunt and Lipo: Chronology, deforestation, and "Collapse

Calibrated Age Ranges

I I I I I I I I I

Ua-11703 _ D [I] - Ua-11702 c.J DD Ua-11701 _ [II] - Ua-11700 ~ Ua-19464 - [I][I] - Ua-19463 [I][I] GrA-2587( f- [I][J] - CNRS- D~ CNRS- f- D~ - SRR-OO c=-J KIA-2038..: i- £J [I] - KIA-1710{ [1][1]

I I I I I I I I I 1000. 1100. 1200. 1300. 1400. 1500. 1600. 1700. 1800. 1900. 2000.

cal AD

Figure 4. Calibrated radiocarbon date (n=12) from Rapa Nui for ancient palm endocarps with cultural a oClatIon by context (archaeological, burned, and/or rat-gnawed). Data compiled from Dran field et al. (1984); Martin on-Wallin and Crockford (2002); Mieth and Bork (2003, 2004); and Orliac (2003).

1) ingle, non-replicated date who e reliability can- lematic given its high-inbuilt age (Taylor and Higham not be demon trated (e.g., potentially from old 1998). This potential must be considered in evaluating the wood, etc.); radiocarbon record. We invite Flenley and Bahn to explain 2) dates from unacceptable material (e.g., coral) and which of these criteria are di pen able in evaluating the samples with mixed i otopic fraction (e.g., reliability and validity of radiocarbon dates, particularly in "bulk" samples of"charcoal and oil mixed'')" or estimating archaeological events uch as colonization of an 3) uncorrected date (i.e., when corrections are not i land. possible; e.g., from marine organi ms, etc., lack­ Regarding our analy i of radiocarbon date (Hunt and ing necessary laboratory analysis), that are known Lipo 2006), Flenley and Bahn (2007b:11) write, "the a ­ to be too old by varying, but unknown, amounts. sembling of nine date from different locations and different For example, Orliac and Orliac (2005:31) recently contexts seems fraught with problems - why hould one reported a ingle date obtained from palm wood charcoal of rea onably expect them to be dating the same thing?" We ca. AD 700. However, dating palm wood is certainly prob- analyze and report eleven pre-750 SP dates, nine from

Rapa Nui Journal 89 Vol. 21 (2) October 2007

Published by Kahualike, 2007 5 Rapa Nui Journal: Journal of the Easter Island Foundation, Vol. 21 [2007], Iss. 2, Art. 2

'Anakena, and two reported from "agricultural sites" (Hunt Ecological Change and the Presence ofHumans and Lipo 2006; see Martins on-Wallin and Crockford 2002 Finally, Flenley and Bahn (2007b: 12) question who compiled a large number of dates). To answer their whether ecological changes (i.e. initial defore tation) will question: we analyze thi et to evaluate an archaeological clo ely mark the time of Polynesian arrival. They ugge t event; the chronology of Rapa Nui colonization. Thu the people might engage in "pre-agricultural activitie uch a different contexts - indeed the more the better - document living on ea bird fi h, sea mammal, etc." Yet in contra­ the same thing: human presence on the island. Because we diction, Flenley and Bahn (2007b: 12) argue that lake-core a k, "When wa Rapa Nui ettled?" dates from across the records, through detection of ecological changes, are more i land compri e the tati tical population. This is identical likely to reveal the earliest signs of human activity than to the many tudie that ha e e tablished chronology in archaeology. The contradiction a sumes that "pre­ ew Zealand (e.g., Ander on 1991; Higham and Hogg agricultural activitie " would not be regi tered in the ar­ 1997; Wilmshurst and Higham 2004) and elsewhere in the chaeological record. The faunal record at 'Anakena shows Pacific (e.g., Spriggs and Anderson 1993; Liston 2005). otherwi e (e.g., Steadman et al. 1994; Hunt 2007). Flenley and Bahn's notions al 0 imply that introduced Polynesian False Assertion ofA Priori Criteria for cultigens were abandoned while people pursued "pre­ Acceptable Dates agricultural" subsistence; and second, introduced rats would Contrary to what Flenley and Bahn (2007b: 11) be­ remain invi ible for hundreds of years. either is biologi­ lieve, we do not exclude early date becau e they form a tail cally pIau ible. Colonist undoubtedly engaged in agricul­ in the di tribution. There i no criterion stipulating exclu- tural activities, a well a hunting, gathering, use of fire, It ion on uch ground. The "lair of the radiocarbon di tri­ and acqui ition of re ource uch as ob idian and ba all. bution (i.e., pre-750 BP dates) is precisely what we analyze. would be nai've to envi ion omething akin to a post­ The notion that a long, cryptic human presence will rtcomparative method cannot claim to picture the hi ­ toncal process." Research in Polyne ia (e.g., Bigg 1972;

Rapa Nui Journal 90 Vol. 21 (2) October 2007 https://kahualike.manoa.hawaii.edu/rnj/vol21/iss2/2 6 Hunt and Lipo: Chronology, deforestation, and "Collapse

and see Terrell et al. 1997) has shown pecific language which also show anomalous dates, may be too relation hip to be far more complex than implied in the impli tic. implistic A to B to C migration models of the 1960 . The common tactic of validating colonization dates (e.g., Emory Given these results and the clear warning, it i diffi­ settling on AD 500 for Rapa ui) based on linguistic mod­ cult to ee how Flenley and Bahn (2007a, 2007b) justify els has langui hed a a mi leading circular argument in electing two dates from a larger pool of demon trably Polynesia (Hunt et al. in pre ). problematic date from bulk sediment sample to make their definitive statement about deforestation and coloniza­ When and Why did Deforestation Occur? tion chronology. Precisely dating hwnan presence on the Flenley and Bahn (2007a, 2007b) cling to dates from island based on the on et of regular fires and vegetation the lake-core samples, particularly those from Rano Kau, by change from pollen requires a dense and continuous edi­ imply a serting that they reliably date the arrival of Poly­ mentary record with [we-grained analyses and multiple ne ian coloni ts on Rapa ui. Yet, as has long been known reliable (i.e., at lea t stratigraphically-consistent) radiocar­ and widely recognized, the radiocarbon results from these bon date on demon trably short-lived specimens (e.g., dat­ sediment core are plagued with problems. ing macro-botanical pecimen using AMS dating and thu Yet, Flenley and Bahn (2007a, 2007b: 12) claim that avoiding high-inbuilt ages, addition of old organic sedi­ deforestation was complete by AD 1000. This stunning ­ ments, etc., but as Butler et al. 2004 show, even these at­ and isolated - declaration puts the completion of deforesta­ tempt may prove problematic given the resolution afforded tion at least 600 year earlier than indications from an im­ by the Rano Kao edimentary record itself). The ideal con­ pre ive corpus of radiocarbon and tratigraphic evidence dition and analyse for lake-core reconstruction of vegeta­ from acro the i land (Arnold et al. 1990' Orliac 2000; tion change do not yet exist for Rapa ui. As Mann et al. Mann et al. 2003, Mieth and Bork 2003, 2004; Mieth et al. (2003: 139, empha i added) correctly pointed out: 2002; Hunt 2007; ee below). This claim also contradicts what Flenley has reported in early and recent publications A close look at Flenley's data show that sedi­ for the same core (e.g. Flenley 1993:43; Flenley and King ments dating to the time ofPolynesian settlement 1984; Butler and Flenley 2001:81). To fit their claim for are either mi ing or highly disturbed in the cores defore tation, Flenley and Bahn (2007a 2007b) 'cherry he analyzed from the crater lake . In hi core pick" dates from an array of inverted and widely di parate from Rano Raraku, ediment ample that were radiocarbon re ults from cores I and 2 at Rano Kao. Draw­ 14C-dated to 480 and 6850 years BP are only 15 ing on result from Core I from Rano Kao Flenley and cm apart in the stratigraphy.... In Rano Kau, Bahn (2007b; see Flenley et al. 1991) appear to settle on ages of 1000 yr BP were obtained on sediments 5 two dates at ca. 1000 BP a ociated with a pollen record for m apart in the core. There are no continuous re­ the loss of forest (but ee Flenley and King [1984:49] where cords of ecological change over the la t 2000 they initially rejected the validity of the 1040±60 BP date, years on Ea ter Island. SRR-2039, given contamination from in-wa hed oil car­ bon). Significantly, Flenley and Bahn's (2007a, 2007b: 12) In a significant 2004 publication, Butler et al. (2004) declaration of virtually complete defore tation by AD 1000 show that dates on various kinds ofsample from Rano Kau is al 0 at dramatic odds with the chronology established by are unreliable, and likely hundreds (even thou and) of three independent re earch teams working on every part of year too old. For example, three dates from samples taken the i land, and publi hing more than 54 radiocarbon dates, from the same upper-most core depth of Rano Kau 2 vary including dates (12) directly on palm endocarp (Figure 3 more than 600 years (Butler et al. 2004:400). In adjacent and 4). These team document a chronology for defore ta­ core depth from Rano Kau 2 bulk ediment plant frag­ tion that con i tently begin after AD 1250-1300, with ign ments, and pollen samples returned ages ranging from more of forest plants persisting into historic times ( ee Figures 3 than 5700 years in age (at only 1.29-1.31 m depth) to tho e and 4; Hunt 2007; Mann et al. 2003; Meith and Bork 2004; below in the same core dating to between ca. 900 BP Orliac 2003). Orliac (2003:192-193), and others (e.g. Kam­ (11.35-11.45 111), ca. 2200 BP (13.40-13.42 m), and ca. 1600 minga and Cotterell 1984), have outlined problems in BP (14.85-14.95 m) (Butler et al. 2004:400). Regarding the Flenley' lake-core chronology for deforestation and the chronology for Rano Kau in particular, Butler et al. attempt to correlate it with cultural decline. Flenley and (2004:395) conclude: Bahn' (2007b) argument for defore tation by AD 1000 makes no en e with regard to their own notions that re­ This serie of 14C age seem to indicate that source depletion led to cultural col lap e, but such both old and young organic components in the "collapse" would have occurred some 600-700 years after sediment are deposited contiguously and that the complete deforestation! Such an argument is at serious odd depositional history of the e core i more com­ with idea that resource depletion led to the abandonment of plex than previously known. Previou age deter­ tatue production and tran port as well. The abundant evi­ mination on bulk sediment from Ea ter I land, dence from multiple field studies does not support Flenley

Rapa Nui Journal 91 Vol. 21 (2) October 2007

Published by Kahualike, 2007 7 Rapa Nui Journal: Journal of the Easter Island Foundation, Vol. 21 [2007], Iss. 2, Art. 2

and Bahn's claim for an early date of deforestation across We wonder what re olved the "open question" a the the island, nor does it fit their story of an ecologically­ historical and ecological evidence for ignificant impact of induced collap e (see Pei er 2005; Rainbird 2002). rats on native vegetation ha only greatly expanded (e.g., Athens et al. 2002; Towns et al. 2006). The Extinct Palms' Lifespan If the extinct palms of Rapa Nui did have a lifespan The que tion of the extinct palm' lifespan is relevant. greater than about 400-500 year the cumulative impacts of However, we do not have an an wer. Flenley and Bahn fire may have finished off the deforestation that rats began (2007a; 2007b) assert that the Jubaea chilensis palm of by depres ing the regeneration of new trees. And Flenley Chile live 2,000 years, but the basi of their speculation is and Bahn (2007b: 12) hould know that rats destroy palm unfounded. Chilean palm expert Juan Grau (2004) refers to seedling (Figure 5) in numerou documented case (e.g., the possibility of a few palms greater than 700 years old on Campbell and Atkin on 1999). In any case, the relative im­ the mainland. While long-lived (i.e., hundreds of years), in pacts of rats, fire, or other effect have not been adequately recent published research on the palms Tomlinson evaluated in the work of Flenley and his associates. Instead, (2006: I0) writes, "the age of the palm can only be deter­ they have precluded te table hypotheses by reducing the mined accurately from knowledge of its seed planting date." i sue to a imp[i tic, unan werable question uch a : "What Extrapolated ages for palms range from 100 to 740 years were they thinking when they cut down the la t palm (Tomlinson 2006: I0). Along these lines, it i noteworthy tree?" (Diamond 1995:68; ee al 0 Bahn and Flenley that the Jubaea chilensis palm planted from seed in the 1992:214). Clearly the synergy of impact from rat as an Temperate Hou e at Kew Garden (England) in 1843, just 164 years ago, i now a large mature palm whose growth is damaging the glass of the greenhouse building. The Rapa ui evidence may suggest a potential lifespan of about 400 years or slightly more, considering the duration of forest survival following human colonization (until about AD 1650; see Orliac 2000), but thi remains only hypothetical. However, using analogie between a living palm and an extinct one may be moot. Botanists have classified the palms that once covered Rapa Nui a Jubaea p. or Pas­ chalococos disperta, de ignating them a unique species or even distinct genus from the native palm on the Chilean mainland. While the data for maximum life pan do not exist, attempt to estimate the extinct palm's life pan across different pecie or genera would remain problematic none­ theless. Moreover, the equable climate of Rapa ui differs ignificantly from the continental environment of Chile further confounding estimate by analogy for palm growth Figure 5. Prehistoric rat-gnawed palm endocarp from Rapa ui rates or life pan. (from the collections in the P. Seba tian Englert Museum); rat­ Flenley and Bahn (2007a, 2007b) spuriously argue gnawed endocarp such a the e provide ideal radiocarbon date that given a 2,000-year life pan for the palm, if rats were for the chronology ofthe extinct palm on Rapa ui. mo tly responsible for deforestation, the palms hould till be on Rapa ui. However, nowhere does Hunt (2006, invasive pecie, direct human action , and perhaps even 2007b) argue that rat were the sole agent of deforestation. climatic variation , make the hi tory of deforestation one ~n tead Hunt raises the question of their relative impact, be t examined in term of its complexity and ecosy tern Just as Flenley and hi colleague did in their original publi­ interaction . cations concerning the i land' defore tation. For example Flenley et al. (1991: 104 emphasis added) wrote: Rats and Other Pacific Islands Flenley and Bahn (2007b:I2) point out that many island have fore t that per i ted despite rat introduction. the effect of introduced rodents on the biota of Two points deserve reiteration here: A long Ii t of island oceanic island are known frequently to have /lot defore ted ays little, ifanything, about Rapa ui. Rapa been disa trous .... and it eems that Easter [s­ Nui is not Fiji, Rarotonga, Tahiti, nor ew Zealand. To ar­ land may have been no exception. Whether the gue a simplistic "rats = deforestation" made on such gener­ extinction of the palm owe more to the preven­ alizations assumes that the diverse island of the Pacific tion of regeneration by rodents, or to the eating have the ame history, biogeography, and ecology. This i of the fruits by man, or to the felling of the ma­ clearly not the ca e. Even a Diamond (1985 :602) pointed ture trees, remains an open question. out many years ago for island biota, "rat have cau ed catastrophic extinction wave on some island, a few ex-

Rapa Nui Journal 92 Vol. 21 (2) October 2007 https://kahualike.manoa.hawaii.edu/rnj/vol21/iss2/2 8 Hunt and Lipo: Chronology, deforestation, and "Collapse

tinctions on others, and no vi ible effect on still others." such a density for Rapa Nui would be unreasonable, as well Second, no one to our knowledge argues that rats (alone) a tell us how many rat bones one should expect to find in succeeded in total deforestation of any i land, although re­ excavations as evidence for fewer rats. search now uggests that perhaps they actually could. Rat Flenley and Bahn ignore the larger faunal record from appear to have devastated the lowland Pritchardia­ 'Anakena (see Hunt 2007: Fig.12), and they fail to appreci­ dominated forests of the Hawaiian Islands, prior to any di­ ate how depo itional factor can result in variable samples rect human impacts of fife or felling (Athens et al. 2002). size of faunal component . The fish bones from 'Anakena Thus the question of the relative impact of rats remains an and other excavations represent primarily in hore taxa open question. Otherwise, it is incumbent upon Flenley and (Martinsson-Wallin and Crockford 2002). Moreover, the Bahn to demonstrate how rats had no impact on the forest presence of sea mammal bones does not require izable of Rapa Nui, and by implication, that humans were the sole canoe and deep- ea fishing, as often cited. In ethnographic agent ofdefore tation. case, dolphin, for example, are taken by using stones struck together in the water to di orient the animals echo­ 'Anakena Evidence location system and drive them into shallow waters, either Flenley and Balm (2007b: 12) write, from small canoes or by people in the water, and then tak­ ing them by hand with little or no specialized technology "interestingly, the excavation at 'Anakena by (e.g., S. Aswani, personal communication 2007; Bloch et al. Steadman et al. (1994) found that the abundance 1990; Porcasi and Fujita 2000; Takegawa 1996). Sea mam­ of rat bones had two peaks at different levels in mal bones from'Anakena (the only location where they are the stratigraphy, about 200 year apart (ca. 1000 reported in any quantity for the island) probably represent BP and ca. 800 BP). Perhaps these could repre­ this specialized capture method, a reported for this location sent the enorrnou plagues hypothesized by in historic times. Deep-sea fishing may have never been a Hunt, although they carcely seem high enough common strategy on Rapa Nui, with or without suitable to do so. But throughout the 200 years, and even trees for canoes. after it, there were abundant fish and dolphin A critical review of the often-repeated claims like bones, suggesting that people were still able to those made by Flenley and Balm for Rapa Nui reveal that go to sea in sizeable canoes. So the rats had ap­ many lack evidence, or that some researchers are too eager parently not succeeded in deforesting the island to settle on simplistic conclusions. in 200 years or more." Violence, Warfare, Cannibalism, and Population While their points made here are perhaps trivial, they Flenley and Balm (2007a:5) pre ent claims about the seem to reflect Flenley and Bahn's misreading or misunder­ nature of violence and warfare in Rapa Nui prehistory and standing of our work. It is worth refuting each to illustrate argue "obsidian spearheads proliferated after the deforesta­ our point: tion, and there i skeletal material that shows severe The age they cite from Steadman as "ca. 1000 BP" wounds" as evidence of "internal warfare" (Flenley and comes from uncorrected marine samples (likely to be at Bahn 2007b: 13). While there was undoubtedly competition least 350 years younger in calibrated age [ca. 650 BP/AD among ancient islanders, there is remarkably little evidence 1300 ] given the marine reservoir effect). They ignore other of "warfare." The primary evidence for it comes from the dates from Steadman's sequence, but select one at "800 oral histories collected late in the post-European contact BP" (note: three dates in this range overlap statistically). history and subsequent to the devastating demographic col­ Furthermore, there is no rationale for treating"1000 BP" as lapse induced by introduced disease and slaving. Archaeo­ a di crete point in time against which to compare "800 BP" logical evidence for pervasive warfare remains ambiguous. and then posit a 200-year interval between the layers that The island's archaeological record boasts no obvious defen­ are the sources of these dated samples. These ages are prob­ sive structures such as the hilltop forts like those found on lematic, and may be statistically the same. Their point re­ Rapa (Kennett et al. 2006) and elsewhere. Use of concealed flects a naive reading of the archaeological and radiocarbon caves as places of refuge may reflect inter-group violence, evidence. but they suggest conflicts on a relatively small-scale. The Nowhere does Hunt (2006, 2007) hypothesize enor­ caves of refuge might also represent places of hiding con­ mous plagues of rats. This is an attempt to sensationalize structed and used as a defense against slave-trading raid th and "controversialize" the issues at hand. Ecological field­ that began in the early 19 Century. work ha documented rat densities of 45-75 rats per acre on Analysis of use-wear in published studies shows that Kure Atoll (Wirtz 1972). At this real-world density, Rapa the so-called "ob idian spearheads" (mota 'a) were tools Nui could have had about 1.9-3.1 million rats within a very (Figure 6) used primarily for activities such as cutting and short period of time (see Fenchel 1974). Given contempo­ scraping plant and animal materials (Church and Rigney rary ecological theory relating population size and growth 1994; Church and Ellis 1996). As Church and Rigney rates in environments with abundant food supplies and no (1994:104) conclude, the predominant use of mata'a was natural predators, Flenley and Bahn should explain why for plant processing (i.e., cutting and scraping green plant

Rapa Nui Journal 93 Vol. 21 (2) October 2007

Published by Kahualike, 2007 9 Rapa Nui Journal: Journal of the Easter Island Foundation, Vol. 21 [2007], Iss. 2, Art. 2

able to violence. The physical evi­ dence suggests that the frequency of warfare and lethal events wa exag­ gerated in folklore." Talk of cannibalism make the tory even more sensational. Yet, even a Bahn (1997) ha pointed out, cannibalism re­ main unproven for Rapa ui. Bahn sug­ ge t there i "no smoke without fire," and that the prevalent ethno-historic dj course on the topic suggests that it had occurred. If cannibalism occurred with orne fre­ quency on the island, we would expect to find evidence in the arcbaeological record, as documented in places such as the American Southwest (e.g., White 1992) and in Fiji (Cochrane et al. 2004). Despite nwnerous excavations and recovery of a large number of hwnan bones, to date, no unambiguous evidence for cannibalism has emerged. We await evidence more ub tan­ Figure 6. Stemmed ob idian artifacts (mota 'a) from the P. Sebastian Englert Museum, tial than tales drawn from the oral tradi­ Rapa ui; these specimens were photographed for documenllition and not selected to tion collected centurie after European represent particular forms. The variability in form is typical and shows that objects po­ contact (see Routledge 1919:212-213). tentially suitable for "weapon" comprise only a very mall fraction ofthe e tools. Shape While the evidence cited for perva ive and use-wear studies indicate that these were multi-purpo e tools that share stems for warfare is weak, it is likely that competi­ hafting. tion for critical resources was a major fac- tor in the evolution of Rapa ui society. parts). They conclude, "tbese activities are inconsistent Competition is widely expressed in ancient Pacific Island with the original proposed function of these tools as spear­ societies, includmg places where defore tation did not oc­ points" (Church and Rigney 1994: 104). Prehistoric artifacts cur. The determmistic equation of deforestation with social of identical form are known from Hawai'i, Tahiti, and else­ and demographic crises cannot account for the many where in Ea t Polynesia, where they are routinely identified counter-example, such as New Zealand or Hawai'i. Re­ as cutting and scraping tool used mainly on plant materi­ search in anthropology and biology reveals that competi­ als. tion is expressed in diverse processes and outcomes. Our inspection of hundreds of these tools from mu­ Curiously, Flenley and Bahn (2007b: 13) write that we seum collection and those from our own field work shows "also pose the question of how the island's population that few, if any, possess traits that would make them effec­ could have risen to crisis proportions if people only arrived tive weapons (Figure 6). While it is conceivable that on in AD 1200." owhere do we raise such a question. De­ rare occasions an ob idjan tool (mata 'a) might have been spite the repeated speculations, we see little, if any, evi­ used to inflict injury, just as today a btchen knife might dence that the i land's population ever reached the large serve as a weapon, there is no evidence for their regular u e number some have cited (e.g., a population of 10-20,000). in inflicting lethal injury. That a European vi itor regarded The archaeological record indicate that ettlement pattern hafted ob idian tools as weapon may reveal more about was disper ed and ub i tence activities extensive. their anxiety, than about the function of the tools. Early As we show above, mathematically, populations could European visitors were mistaken about other things as well. have grown to large sizes even with a AD 1200 coloniza­ Skeletal material from Rapa ui provides little in the tion time frame (Figure 2; see Birdsell 1957). However, way of support for widespread warfare or violence. Pub­ there is no evidence that the Rapanui population reached lished studie of hundreds of skeletons show that evidence the huge number cited. Speculations about a large popula­ for violent injury or fatalities is minimal. Based on a study tion do not support a longer chronology, regardle of the of2 618 human bone, Ow ley et al. (1994: 164) report that predilections of orne to make circular argwnent . only 2.5% of crania show antemortem fractures with evi­ dence of healing or perimortem breakage cau ed by trau­ Co CLUSIO matic injuries. Owsley et al. (1994: 174) conclude, The dramatic tory of Rapa Nui's so-called "eeocide" wa "most skeletal injuries appear to have been constructed with fajtb in a long chronology, peculations of nonlethal. Few fatalities were directly attribut- a huge prehi toric population ize, and a puriou Iy dated

Rapa Nui Journal 94 Vol. 21 (2) October 2007 https://kahualike.manoa.hawaii.edu/rnj/vol21/iss2/2 10 Hunt and Lipo: Chronology, deforestation, and "Collapse

lake core record ofdeforestation for the island. Our ongoing more than political rhetoric. What happened in the past oc­ field research, the recent impressive palaeo-environmental curred regardless of the political framing imposed by some work of several teams, as well as comparative research re­ today. We believe the issues should be resolved with histori­ cently published for the Hawaiian Islands, has changed cal evidence and the empirical standards of modern archae­ some perspectives and raises new questions about Rapa ology. Let the political implications follow from our re­ Nui's historical ecology. In Hunt's (2006, 2007) recent re­ search, rather than allow politics to dictate our findings. views he has examined archaeological, palaeo­ It does not matter whether Rapa Nui offers a parable environmental, and contemporary ecological evidence in for today's urgent environmental problems. They are urgent consideration of a significant role for the Pacific rat in Rapa problems nonetheless. Nowhere do we even hint that the Nui's ecological catastrophe. The fact that rats alone can current global environmental crisis should be ignored. Our devastate forests raises the issue of their relative impacts for concern echoes Grayson and Meltzer (2003:590), that sci­ Rapa Nui, as well as for other island ecosystems. The role ence and critical environmental issues are both done a dis­ of rats has often been underestimated, yet this does not deny service by relying on accounts with virtually no empirical that direct human actions such as the use of fue likely support. As Hunt (2006:419) remarked, "mistakes or exag­ played decisive roles in deforestation. Additional research gerations only lead to oversimplified answers and hurt the should disentangle the relative impacts of contributing fac­ cause of environmentalism. We will end up wondering why tors. In short, the environmental catastrophe of Rapa Nui our simple answers were not enough to a make a difference likely has a complex history, one that has been obfuscated in confronting today's problems." by simple speculations on the intentions of the person cut­ Finally, it is worth reiterating that science is not based ting down the last tree. As the story of rats as invasive spe­ on belief, faith, or assertion, but on the hard evidence we cies suggests, perhaps the "last tree" simply died, and rats acquire in attempts to prove ourselves wrong. We invite ate the last seeds (Hunt 2007:499). We argue tllat Rapa Nui Flenley and Bahn to join us in evaluating the complexity of may tell of the effects of invasive species, invasional melt­ Rapa Nui's archaeological, ecological and evolutionary down (Simberloff and Von Holle 1999), and the synergy of history. In the framework of science, we challenge them to these introduced element when they reach evolutionary open their minds to the emerging picture of Rapa Nui's re­ isolates in the remote islands of the Pacific. markable prehistory, even if it is sometimes different than Flenley and Bahn (e.g., 2002, 2007b) have convinced all of us have supposed. themselves that Rapa Nui society crashed before Europeans arrived in 1722. Apparently in support of Rapa Nui as a Address correspondence to: [email protected] scary parable for our own environmental woes, they believe deforestation caused popuJation collapse. Diamond (1995, 2005) has profited from telling the same story. Yet, this as­ REFERENCES sumption conflates the undisputed fact ofdeforestation with the specuJation that the population was much larger, but Anderson, A. L991. The chronology ofcolonization in New Zea­ then collapsed before European contact. Stated simply, no land. Antiquity 65:767-795. Ander on, A. 1995. Current approaches in East Polynesian coloni­ evidence exists for a pre-European contact population col­ zation research. Journal ofthe Polynesian Society 104 lapse. It is merely supposition, repeated over and over. A (1):110-132. pre-contact demographic collapse remains untested, undem­ Anderson, A. and Y. Sinoto. 2002. New radiocarbon ages ofcolo­ onstrated archaeologically. The historic slave-trading, epi­ nization sites in East Polyne ia. Asian Perspectives 41 :242­ demic disea es, intensive sheep ranching, and tragic popula­ 257. tion collapse - indeed the genocide of the Rapanui People ­ Arnold, M., M. Orliac and H. Valladas. L990. Donnees Nouvelles is well documented, and has been recognized for a long sur La Disparition du Palmier (cf. Jubaea) de I'I1e de Paques. time (e.g., Metraux 1957). It seems that Flenley and Bahn Courier Forschungsinslilut Senckenberg 125:217-219. (2007a, 2007b) are unaware of both the historic impacts on Athens, J. S. 1997. Hawaiian native lowland vegetation in prehis­ tory. Historical Ecology in the Pacific Islands: Prehistoric Rapa Nui as well as the significant literature on the biologi­ Environmental and Landscape Change. P. V. Kirch and T. L. cal impacts Europeans wrought in the Americas and the Hunt, eds.:248-270. New Haven: Yale University Press. Pacific. As Peiser (2005) has pointed out, some writers have Athen ,J. S., H. D. Tuggle, J. V. Ward and D. 1. Welch. 2002. transposed an actual documented historic genocide with Avifaunal extinctions, vegetation change, and Polynesian prehistoric "ecocide" where the victims are blamed for their impacts in prehistoric Hawai'i. Archaeology in Oceania own demise. 37:57-78. Flenley and Balm (2007b) suggest that questioning the Bahn, P.G. 1997. Easterlsland or (Man-) Eaters Island? Rapa Nui impacts of European contact reflects "anti-European bia ," Journal 11(3):123-125. or a "Euro-phobic" model. We find such attributions reveal­ Bahn, P. G. and J. Flenley. 1992. Easter Island, Earth Island. New York: Thames and Hudson. ing and troubling. Following such logic, are Flenley and Barnes, S., L. Matisoo-Smith and T. Hunt. 2006. Ancient DNA of Balm therefore, anti-Polynesian or Euro-centric? Unfortu­ the Pacific rat (Rattus exulans) from Rapa Nui (Easter Is­ nately, such notions inevitably obscure the generation of land). Journal ofArchaeological Science, 33: 1536-1540. scientific knowledge and reduce historical questions to little

Rapa Nui Journal 95 Vol. 21 (2) October 2007

Published by Kahualike, 2007 11 Rapa Nui Journal: Journal of the Easter Island Foundation, Vol. 21 [2007], Iss. 2, Art. 2

Bergsland, K. and H. Vogt. 1962. On the validity ofglottochronol­ Dunnell, R. C. and M. Readhead. 198 . The relation ofdating and ogy. Current Anthropology 3: 115-153. chronology: Comment on hatters and Hoover (1986) and Bigg ,B. 1972. Implication oflingui tic ub-grouping with pe­ Butler and Stein (1988). QuatemGlY Re earch 30:232-233. cial reference to Polynesia. Studies in Oceanic Culture His­ Emory, K.P. 1946. Ea tern Polynesia: Its cui/ural rela/ionships. tory. R. Green and M. Kelly ed .:143-160. Vol. 3. Bernice P. Ph.D. dis ertation, Yale University. Bi hop Museum, Honolulu. Emory, K. P. 1963. East Polynesian relation hip: Settlement pat­ Birdsell, J. B. 1957. Some population problems involving Pleisto­ tern and time involved as indicated by vocabulary agree­ cene man. Cold Spring Harbor Sympo ia on Quantitative ments. Journal ofthe Polynesian Society 72:78-100. Biology 22:47-69. Fenchel, T. 1974. Intrinsic rate ofnatural increase: The relation­ Bloch, D., G. Desportes, K. Hoydal and P. Jean. Pilot whaling in ship with body ize. Oecologia 14(4):317-326. the Faroe Islands, July 1986-July 1988. North Atlantic Stud­ Flenley, J. R. 1993. The palaeoecology of Rapa Nui and its eco­ ies 2 (1,2):36-44. logical di a ter. Easter Island Studie . Contribu/ions /0 the Bloomfield, L. 1933. Language. Chicago: University ofChicago. His/ory ofRapanui in MemOlY ofWilliam T. Mulloy. S. R. Burney, D. A., H. F. Jame L. P. Burney, . L. 01 on, W. Kiku­ Fischer, ed:27-45.0xford: Oxford Book. chi, W. Wagner, M. Burney, D. Mc loskey, D. Kikuchi, F. Flenley, J. R. and P. Bahn. 2002. The Enigmas ofEa fer lland. V. Grady, R. Gage and R. i hek. 200 I. Fossi I evidence for a New York: Oxford Univer ity Pre s. diver e biota from Kaua'i and its transformation ince human Flenley, J. R. and P. Bahn. 2007a. Ratted out. American Scien/is/ arrival. Ecological Monographs 71(4):615-641. 95( I):4-5. Burney, L. P. and D. A. Burney. 2003. Charcoal stratigrapbies for Flenley, J. R. and P. Bahn. 2007b. Conflicting views of Ea ter Kaua'i and the timing ofhuman arrival. Pacific Science 57 Island. Rapa Nui Journal 21 (I): I 1-13. (2):211-226. Flenley, J. R. and S. King. 1984. Late Quaternary pollen records Butler, K. and J. Flenley. 200 I. Further pollen evidence [rom from Easter Island. Nature 307:47-50. Easter Island Proceedings ofthe Fifth International Confer­ Flenley, J. R., S. King, J. Jackson, C. Chew, J. Teller and M. Pren­ ence on Easter Island and the Pacific. C. teven on, G. Lee tice. 1991. The late Quaternary vegetational and climatic and F. J. Morin, eds.:79-86. Los Osos: Easter I land Founda­ hi tory of Easter Island. Journal ofQua/ernary Science 6:85­ tion. 115. Butler, K., C. A. Prior and J. Flenley. 2004. Anomalou radiocar­ Gibbon, A. 2006. Date revi e Ea ter I land hi tory. Science bon dates from Easter Island. Radiocarbon 46(1):395-405. 311:1360. Campbell, D. J. 1978. The effect ofrat on vegetation. The Ecol­ Grau, J. 200 l. More about Jubaea chilensis on Ea ter I land Pro­ ogy and Control ofRodents in New Zealand ature Reserves. ceedings ofthe Fifth International Conference on Easter P. R. Dingwall, I. A. E. Atkinson and C. Hay, eds. 4:99-120. Island and the Pacific. C. tevenson, G. Lee and F. J. Morin, Wellington: Department of Land and Survey Information ed .:87-90. Los Osos: Easter Island Foundation. Serie . Grau, J. 2004. Palmeras de Chile. Santiago: Oikos. Campbell, D. J. and I. A. E. Atkin on. 1999. Effects of kiore Grayson, D. K. and D. J. Meltzer. 2003. A requiem for orth (Rattus exulans Peale) on recruitment of indigenou coastal American overkill. Journal ofArchaeological Science trees on northern offshore i land of ew Zealand. Journal of 30:585-593. the Royal Society ofNew Zealand 29(4):265-290. Green, R. C. 1966. Linguistic ub-grouping within Polynesia: The Campbell, D. J. and I. A. E. Atkinson. 2002. Depre ion oftree implication for prehi toric settlement. Journal of/he Polyne­ recruitment by the Pacific rat (Rattus exulans Peale) on ew sian Society 75:6-38. Zealand's northern offshore islands. Biological Conservation Heyerdahl, T. 1961 a. An introduction to Ea ter Island. Archae­ I07( I): 19-35. ology ofEaster Island, T. Heyerdahl and E. . Ferdon, cd . Church, F. and J.G. Ellis. 1996. A use-wear analysis ofobsidian :21-90. Monographs ofTbe chool of American Re earch tool from an ana kionga. Rapa Nui Journal 10(4):81-88. and the Museum of ew Mexico, 24(1). London: George Church, F. and J. Rigney. 1994. A micro-wear analysis oftools Allen and Unwin. from Site 10-241, Easter Island - an inland proce sing site. Heyerdahl, T. 1961 b. Conclusion. Archaeology ofEas/er Island, Rapa Nui Journal 8(4): I0 I-I 05. T. Heyerdahl and E. . Ferdon, eds.:493-526. Monographs of Cochrane, E. E., M. Pietrusewsky and M.T. Douglas. 2004. Cul­ The School ofAmerican Research and the Museum ofNew turally modified human remains recovered from an earth­ Mexico, 24(1). London: George Allen and Unwin. oven interment on Waya 1 land, Fiji. Archaeology in Oceania Higham, T. F. G. and A. G. Hogg. 1997. Evidence for late Polyne­ 39( 1):54-59. sian colonization of ew Zealand: Univer ity ofWaikato Dean, J. . 1978. Independent dating in archaeological analy is. Radiocarbon mea urement . Radiocarbon 39(2): 149-192. Advances in Archaeological Method and TheOlY 1:223-255. Hunt, T. L. 2006. Rethinking the fall ofEaster Island: ewevi­ Denham T., F. J. Eble, W. and J. V. Ward. 1999. Palaeoenviron­ dence points to an alternative explanation for a civilization' mental and archaeological inve tigations at Ohi'apilo Pond, collapse. American Scientist 94:412-419. leeward coast of Moloka'i, Hawai'i. Hawaiian Archaeology Hunt, T. L. 2007b. Rethinking Easter Island's ecological cata tro­ 7:35-39. phe. Journal ofArcheological Science 34:485-502. Diamond, J.1985. Rats a agent ofextermination. Nature 31 Hunt, T. L. and C. P. Lipo. 2006. Late colonization of Easter I­ 602-603. land. Science 311: 1603-1606. Diamond, J. 1995. Ea ter end. Di cover 9:62-69. Hunt, T. L., C. P. Lipo, and M. Mad en. In pre s. The flaws of Diamond, J. 2005. Collap e: How Societies Choo e to Fail or phylogeny as hi tory: The Pacific Islands case. Journal of Succeed. New York: Viking. Evolu/ionary and Historical Sciences. Dransfield, J., J. R. Flenley, S. M. King, D. D. Harkness and S. Kamminga, J. and 8. Cotterell. 1984. The pollen record and Easter Rapu. 1984. A recently extinct palm from Easter Island. a­ Island tatues. Nature 312:289. ture 312(5996):750-752.

Rapa Nui Journal 96 Vol. 21 (2) October 2007 https://kahualike.manoa.hawaii.edu/rnj/vol21/iss2/2 12 Hunt and Lipo: Chronology, deforestation, and "Collapse

Kennett, D., A. Anderson, M. Prebble, E. Conte and J. Southon. Pei er, B. 2005. From genocide to ecocide: the rape of Rapa ul. 2006. Prehistoric human impact on Rapa, French Polynesia. Energy and Environment 16 (3 and 4):513-539. Antiquity 80: 1-15. Po rca i, J. F. and H. Fujita. 2000. The dolphin hunter: A special­ Kirch, P. V. 1997. Microcosmic hi tories: Island perspectives on ized prehistoric marine adaptation in the southern California "global" change. American Anthropologist 99:30-42. Channel Islands and Baja California. American Antiquity 65 Kirch, P. V. 2004. Oceanic Islands: microcosms of"global" (3):543-566. change. The Archaeology o/Global Change: The impact 0/ Rainbird, P. 2002. A me sage for our future? The Rapa ui humans on their environment. C. L. Redman, S. R. Jame , P. (Easter Island) eco-disaster and Pacific I land environment. Fish and J. D. Rogers, eds.:13-27. Smithsonian, Wa hington World Archaeology 33:436-451. DC. Routledge, K. 1919. The MystelY 0/Easter Island. London: ifton, Liston, J. 2005. An assessment of radiocarbon dates from Palau, Praed. we tern Micronesia. Radiocarbon 47:295-354. Simberloff, D. and B. Von Holle. 1999. Positive interactions of Lovgren, S. 2006. Easter I land settled later, depleted quicker than non-indigenous species: invasional meltdown? Biological tbought? National Geographic eli's (on line), March 9, Invasions I(I ):21-32. 2006. Skj01 voId, A. I994a. Archaeological investigations at Anakena, Mann, D., J. Chase, J. Edwards, R. Beck, R. Reanier and M. Ma Ea ter I land Archaeological Investigations at Anakena, 2003. Prehi toric de truction of the primeval soils and vege­ Easter Island. A. Skj0lsvold, ed.:5-120. vol. 3. The Kon Tiki tation ofRapa ui (Isla de Pa cua, Easter Island). Easter Museum Occa ional Papers. Oslo: The Kon Tiki Mu eum. Island: Scientific Exploration in the Worlcl's Environmental Skj01 void, A. I994b. Appendix: Osteological analysis. Archaeo­ Problems in Microco m. J. Loret and J. T. Tanacredi, logical Investigations at Anakena, Easter 1 land. A. Skj0ls­ eds.: 133 -153. New York: Kluwer AcademicfPlenum. void, ed. :203-211. Vol. 3. The Kon Tiki Museum Occa ional Martin on-Wallin, H., and S. J. Crockford. 2002. Early settlement Paper. Oslo: The Kon Tiki Mu eum. ofRapa ui (Easter Island). Asian Perspectives 40(2):244­ Skotlsberg, C. 1956. The atural History 0/Juan Femande= and 278. Easter Island. Vol.l. Upp ala: Almsquist and Wiksells. Metraux, A. 1957. Ea tel' iland: A Stone-Age Civiliz(ltion o/the Sprigg, M. and A. Ander on. 1993. Late colonization ofEa t Pacific. London: Andre Deut ch. Polyne ia. Antiquity 67:200-217. Mieth, A. and H.-R. Bork. 2003. Diminution and degradation of Steadman, D. W. 2006. Extinction and Biogeography o/Tropical environmental resources by prehistoric land use on Poike Pacific Birds. Chicago: University ofChicago Pre . Penin ula, Easter I land (Rapa ui). Rapa Nui Journal 17 teadman, D. W., P. Varga and C. Cri tino. 1994. tratigraphy, (I ):34-42. chronology, and cultural context ofan early faunal a em­ Mieth, A. and H.-R. Bork. 2004. Easter Island - Rapa Nui: Scien­ blage from Easter I land. Asian Perspectives 33(1 ):79-96. tific Pathways to Secrets o/the Past. Man and Environment Takegawa, D. 1996. The method ofdolphin hunting and the di tri­ I. Department ofEcotechnology and Eco ystem Develop­ bution of teeth and meat: Dolphin hunting in the olomon ment. Kiel, Germany: Ecology Center, Chri tian-Albrecht ­ I land 2. Coastal Foragers in Transition. T. Akimichi, ed. Universitat zu Kiel. :67-80. Semi Ethnological Studie No. 42. Osaka: National Mieth, A. and H.-R. Bork. 2005. Traces in the soils: interaction Museum of Ethnology. between environmental change, land u e and culture in the Taylor, J. E. and T. F. G. Higham. 1998. Problems ofQumran's (pre)history ofRapa ui (Ea ter [ land). The Reiiaca Papers. chronology and the radiocarbon dating of palm log sample VI International Conference on Rapa Nui and the Pacific. C. in locu 86. The Qumran Chronicle 8( 1/2):83-94. Steven on, J. M. Ramirez, F. J. Morin and . Barbacci, Terrell, J., T. L. Hunt, and C. Go den. 1997. The dimensions of eds.:55-65. Lo 0 0 : Easter [ land Foundation. social life in the Pacific: Human diver ity and the myth of the Mieth, A., I-l-R. Bork and I. Feeser_ 2002. Prehistoric and recent primitive isolate. Current Anthropology 38(2): 155-195. land use effects on Poike Peninsula, Easter Island (Rapa Nui). Tomlinson, P. B. 2006. The uniqueness of palms. Botanical Jour­ Rapa Nui Journal 16(2):89-95. nal o/the Linnean Society 151 :5-14. Mulloy, W. and G. Figueroa. 1978. The A Kivi- Vai Teka Complex Towns, D. R., I. A. E. Atkin on and C. H. Daugherty. 2006. Have and its Relation hip to Easter I land Architectural Prehi ­ the harmful effects of introduced rats on island been exag­ tory. Asian and Pacific Archaeology Series 8. Social Science gerated? Biological Invasions 8:863-891. Research Institute, Honolulu: Univer ity ofHawaii. White, T. 1992. Prehistoric Cannibalism at Mancos 5MTUMR­ Orliac, C. 2000. The woody vegetation ofEa ter Island between 2346. Princeton: Princeton University Pre . the early 14th and the mid-17th Centuries A.D. Easter Island Wilmshurst, J. M. and T. F. G. Higham. 2004. U ing rat-gnawed Archaeology: Research on Early Rapanui Culture. C. Steven­ seeds to independently date the arrival of Pacific rat and on and W. Ayre , ed .: J99-207. Los 0 os: Easter r land humans in New Zealand. The Holocene 14(6):801-806. Foundation. Wirtz, W. O. 1972. Population ecology ofthe Polyne ian rat, Rat­ Orliac, C. 2003. Ligneux et palmiers de l'i1e de Paques du Xleme tus exulans, on Kure Atoll, Hawaii. Pacific Science 26:433­ au XVIleme siecle de notre ere. Archeologie en Oceanie 464. insulaire: Peuplement, societe et paysage . C. Orliac, Young, E. 2006. A monumental collap e. New Scientist 29 :30-34. ed.: 184-199. Pari: Editions Artcom. Orliac, C. and M. Orliac. 2005. La flore di parue de l'i1e de Pa­ que. Les nouvelles de I'archeologie 102(4):29-33. Owsley, D. W., G. Gill, and S. D. Ousley. 1994. Biological effect ofEuropean contact on Ea ter Island. In the Wake o/Contact: Biological Responses to Conquest, C. S. Lar en and G. R. Milner,eds.:161-177. ewYork: Wiley-Liss, Inc.

Rapa Nui Journal 97 Vol. 21 (2) October 2007

Published by Kahualike, 2007 13